A DCS intelligent control system and optimization method for multi-product co-production

Abstract

The application relates to the field of DCS intelligent control, and discloses a DCS intelligent control system and optimization method for multi-product co-production, which comprises the following steps: collecting process parameters, material flow parameters, energy flow parameters, emission parameters and equipment state parameters corresponding to each product node in the multi-product co-production process, and constructing a state coupling model; based on the state coupling model, constructing an untouchable bottom line constraint set containing safety constraints, emission constraints, equipment limit constraints and system stability constraints, and forming a feasible state space; calculating a node comprehensive gain and loss change rate caused by system state change and a safety margin change rate relative to the bottom line constraint; generating a plurality of state candidate adjustment direction sets, forming an executable path for gradually adjusting from a current state to a target state; and mapping to adjustment instructions for each product node and equipment unit, and generating a state adjustment record. The application has the advantage of improving operation stability.

Description

Technical Field

[0001] This invention relates to the field of DCS intelligent control, specifically to a DCS intelligent control system and optimization method for multi-product joint production. Background Technology

[0002] As process industries move towards higher efficiency, lower energy consumption, and cleaner production, multi-product co-production models are widely used to improve resource utilization. In this model, different products share raw materials, equipment, and utilities, resulting in significant complexity and parameter coupling in the production process. During multi-product co-production, process parameters are highly correlated; changes in a single parameter often simultaneously affect output, quality, energy consumption, and side reactions, exhibiting strong coupling characteristics. However, existing control systems primarily employ single-parameter, single-loop control methods, making it difficult to effectively manage the mutual influence between multiple parameters. Because control remains at the local parameter level, existing technologies generally struggle to coordinate conflicting relationships between different production objectives. For example, increasing output often leads to increased energy consumption, quality fluctuations, or increased byproducts. Furthermore, pollutant emissions mainly rely on end-of-pipe treatment, making source control difficult. Under multi-product co-production conditions, this reactive approach results in a delayed response, easily leading to resource waste and environmental risks. Therefore, it can be seen that existing multi-product co-production control systems remain at the single-parameter, local-loop level, lacking system-level modeling and collaborative control capabilities for strongly coupled production processes, making it difficult to achieve unified optimization of multiple objectives. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a DCS intelligent control system and optimization method for multi-product joint production, which has the advantage of improving operational stability and solves the problems mentioned in the background technology.

[0004] To achieve the aforementioned goal of improving operational stability, this invention provides the following technical solution: a DCS intelligent control optimization method for multi-product co-production, comprising the following steps: Collect process parameters, material flow parameters, energy flow parameters, emission parameters and equipment status parameters corresponding to each product node in the multi-product co-production process, and construct a state coupling model of the mutual influence relationship between the states of each node by combining the raw material sharing relationship, equipment coupling relationship and utility constraint relationship between products. Based on the state coupling model, a set of inviolable bottom line constraints is constructed, including safety constraints, emission constraints, equipment limit constraints and system stability constraints. System states that violate any bottom line constraint are eliminated to form a feasible state space. Within the feasible state space, a joint analysis is performed on the changes in output, energy consumption, emissions, and equipment load at each node to calculate the rate of change of the overall gains and losses of the nodes caused by the changes in system state and the rate of change of the safety margin relative to the bottom line constraint. Based on the overall gain and loss rate of nodes and the safety margin rate of change, multiple sets of candidate adjustment directions for the current state are generated. By comparing the relative changes of the overall gain and loss rate of nodes and the safety margin rate of change in each candidate direction, an executable path is formed to gradually adjust from the current state to the target state. The executable path is mapped to adjustment instructions for each product node and equipment unit, and the adjustment instructions are executed through DCS to generate status adjustment records.

[0005] Preferably, the process of constructing a state coupling model of the mutual influence relationships between the states of each node is as follows: The process parameters, material flow parameters, energy flow parameters, emission parameters, and equipment status parameters collected at each product node are normalized. Calculate the coupling weight matrix between nodes by combining the raw material sharing relationship, equipment coupling relationship and utility constraint relationship between products; A multidimensional state vector is constructed based on the node coupling weight matrix to form a state coupling model in which each node influences the others.

[0006] Preferably, the process of constructing a set of inviolable bottom-line constraints, including safety constraints, emission constraints, equipment limit constraints, and system stability constraints, is as follows: Based on the state coupling model, the parameters of safety constraints, emission constraints, equipment limit constraints and system stability constraints are quantified and represented to form a constraint vector; The constraint vectors are classified and graded to generate a baseline constraint matrix based on the node, device, and system levels; Output constraint matrix file, combining the constraint parameters to form a set of untouchable bottom line constraints.

[0007] Preferably, the process of forming a feasible state space is as follows: The possible states of each node in the state-coupled model are compared and verified with the set of untouchable bottom line constraints. States that violate any constraint are eliminated, and a preliminary set of feasible states is output. For each feasible state marked as a tightly constrained node, output the node constraint identification information; The initial set of feasible states is integrated with the constraint identifiers to generate the final feasible state space.

[0008] The preferred method for jointly analyzing changes in output, energy consumption, emissions, and equipment load at each node is as follows: Within the feasible state space, for each feasible state, calculate the node output, energy consumption, emissions, and equipment load indicators, and output a set of node state indicators. Analyze the mutual influence, changing trends, and sensitivity between nodes, generate node dependency information, and output a dependency table; The node status indicators and information in the dependency table are classified and processed to form a node change classification result; The node status indicators and classification results are standardized to output standardized node data.

[0009] Preferably, the process of calculating the rate of change of the overall gains and losses of nodes caused by changes in system state and the rate of change of the safety margin relative to the bottom-line constraint is as follows: Calculate the overall gain and loss change rate of nodes based on standardized node data, and output a gain and loss change result table; Calculate the rate of change of node safety margin using the list of bottom line constraints, and output the safety margin result table; Integrate the gain / loss change results table with the safety margin results table to output a dual change rate evaluation table; The dual rate of change evaluation table is sorted and normalized, and the sorting result table is output.

[0010] Preferably, the process of generating multiple sets of candidate state adjustment directions is as follows: Multiple candidate adjustment directions are extracted from the sorting result table. Each direction corresponds to a specific combination of node changes, forming a preliminary set of candidate directions. After eliminating the initial candidate directions that result in a decrease in overall gains and losses and a reduction in safety margin, a set of candidate directions is obtained after screening. Generate a priority index for the filtered candidate direction set and output a direction priority list; Based on the priority list, each direction is mapped to a specific node state combination path, generating multiple sets of candidate adjustment directions for the state.

[0011] Preferably, the process of forming an executable path that gradually adjusts from the current state to the target state is as follows: Select a direction from a set of multiple candidate state adjustment directions, and map the node combination changes corresponding to each direction into a node state adjustment sequence to form a preliminary executable path; Perform bottom-line constraint verification on the status of each node in the preliminary executable path and generate path verification results; The sequence of node states that pass the verification constitutes the final executable path.

[0012] Preferably, the process of generating state adjustment records is as follows: The state of each node in the executable path and its corresponding adjustment instructions are converted into structured records. Record the output, energy consumption, emissions, and equipment load status of each node during execution, and output a node execution status table; The results of the adjustment command execution are integrated with the path verification information to generate a complete status adjustment record.

[0013] A DCS intelligent control and optimization system for multi-product co-production includes: State modeling module: Based on the process parameters, material flow, energy flow, emissions and equipment status of each node, and combined with the constraints of raw material sharing, equipment coupling and public works, a state coupling model of mutual influence of each node is constructed. Constraint Analysis Module: Utilizes a state-coupled model to generate a set of constraints for safety, emissions, equipment limits, and system stability, filters feasible state spaces, and outputs feasible node states; Direction generation module: Calculates the overall gains and losses and safety margin changes of nodes in the state space of feasible nodes, generates multiple state candidate adjustment directions and sorts them by priority; Path planning module: Based on the set of candidate adjustment directions, it forms an executable path that is gradually adjusted from the current state to the target state, and outputs the path verification results; Instruction execution module: Maps executable paths to node status adjustment instructions, executes them through DCS, and generates complete status adjustment records.

[0014] Compared with the prior art, the present invention provides a DCS intelligent control system and optimization method for multi-product joint production, which has the following beneficial effects: This invention achieves global collaborative perception and constraint management of the states of each product node and equipment unit by constructing a refined state coupling model and an inviolable bottom-line constraint set. Under the premise of ensuring safety, emission compliance and stable equipment operation, the system can dynamically evaluate the comprehensive gains and losses of nodes and changes in safety margins, generate multi-directional adjustment schemes and plan the optimal execution path, thereby significantly improving the flexibility of the production process, energy utilization efficiency, emission control level and overall operational stability. At the same time, the method directly maps the executable path to DCS adjustment commands and generates complete state adjustment records, realizing operation automation, traceability and system stability, reducing the risk of human intervention, and providing data support for subsequent process optimization and decision-making, thereby improving the economic benefits, safety, operational reliability and sustainable operation capability of the multi-product co-production system. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the method of the present invention; Figure 2 This is a schematic diagram of the structure of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, those skilled in the art who have not made any innovative embodiments are all within the scope of protection of the present invention.

[0017] Example 1: Please refer to Figure 1 As shown in the embodiment of the present invention, a DCS intelligent control optimization method for multi-product joint production includes the following steps: S1: Collect process parameters, material flow parameters, energy flow parameters, emission parameters and equipment status parameters corresponding to each product node in the multi-product co-production process. Combine the raw material sharing relationship, equipment coupling relationship and utility constraint relationship between products to construct a state coupling model of the mutual influence relationship of each node status.

[0018] The process of constructing a state coupling model in S1 to demonstrate the mutual influence between the states of each node is as follows: The process parameters, material flow parameters, energy flow parameters, emission parameters, and equipment status parameters collected at each product node are normalized. Each type of parameter is synchronized according to the time series acquisition frequency, formatted into a unified data structure, such as stored in a table or time series vector, and missing values ​​are filled and outliers are removed to ensure data integrity and accuracy. Then, each parameter is normalized to map parameters of different dimensions and magnitudes to a fixed numerical range, such as [0,1], to eliminate the influence of dimension differences on subsequent calculations, so that coupled calculations can be performed under a unified standard, and to ensure that the comparison and weighting of the states of each node are operable and consistent.

[0019] Calculate the coupling weight matrix between nodes by combining the raw material sharing relationship, equipment coupling relationship and utility constraint relationship between products; The raw material sharing, energy flow dependence, and equipment sharing relationships among nodes are analyzed to construct a node relationship table to clarify the scope of each node's dependence on other nodes. Then, a coupling weight matrix is ​​formed through weighted calculation, where the weights reflect the intensity of the impact of node state changes on related nodes. The weight calculation can be based on the proportion of shared raw materials, the degree of equipment load dependence, and flow sensitivity. During the matrix construction process, coupling response delay and coupling differences between nodes are considered simultaneously, and the weights are standardized to ensure that the coupling relationships can be directly used for multidimensional state vector construction under a unified scale.

[0020] A multidimensional state vector is constructed based on the node coupling weight matrix to form a state coupling model in which each node influences the other. By using the inter-node coupling weight matrix, the normalized parameters of each node are combined into a multi-dimensional state vector. Each dimension of the vector corresponds to the node's process parameters, material flow, energy flow, emissions, and equipment status information. Coupling weights are introduced into the vector to reflect the mutual influence between nodes. The state vectors of all nodes are integrated according to the coupling relationship to form a complete state coupling model of the interaction between the states of each node.

[0021] S2: Based on the state coupling model, construct a set of untouchable bottom line constraints that includes safety constraints, emission constraints, equipment limit constraints and system stability constraints, and eliminate system states that violate any bottom line constraint to form a feasible state space.

[0022] The process of constructing the set of untouchable bottom line constraints in S2, which includes safety constraints, emission constraints, equipment limit constraints, and system stability constraints, is as follows: Based on the state coupling model, the parameters of safety constraints, emission constraints, equipment limit constraints and system stability constraints are quantified and represented to form a constraint vector; After completing the state coupling model construction, the constraint values ​​of each node and equipment are obtained from the equipment manual, process specifications and environmental emission standards, such as safety pressure, upper temperature limit, emission limit and system stability index. Then these constraint parameters are converted into a calculable numerical form to form a constraint vector corresponding to each constraint. The vector elements include constraint type, constraint limit, influencing node and weight information. In the quantification process, the coupling relationship between nodes also needs to be considered to ensure that the constraint parameters can reflect the potential impact of changes in a single node on the entire system.

[0023] The constraint vectors are classified and graded to generate a baseline constraint matrix based on the node, device, and system levels; According to the constraint type, the vectors are divided into four categories: safety constraints, emission constraints, equipment limit constraints, and system stability constraints. Each type of constraint occupies an independent column or block in the matrix. The constraints are classified according to their importance, scope of influence, and node coupling strength, generating node-level, equipment-level, and system-level constraint hierarchy information. The classified constraints are arranged in matrix form, with the matrix rows representing nodes or equipment and the columns representing constraint types and constraint limits. The matrix elements contain constraint values, weights, and priority information. This matrix can be directly used for subsequent feasible state screening and constraint verification.

[0024] Output constraint matrix file, combining each constraint parameter to form a set of untouchable bottom line constraints; Based on the constraint limits and weight information of each node and device in the matrix, a set of constraint rules for all possible states of the system nodes is generated and saved in the form of a data file, including constraint type, constraint threshold, constrained node and dependency relationship. During the combination process, logical verification is performed on possible constraint conflicts or cross-effects to ensure that each constraint rule can be accurately identified during system execution. The final set of untouchable bottom line constraints can be directly used as input for feasible state space filtering, realizing global constraint management and real-time verification of system state changes.

[0025] The process of forming a feasible state space in S2 is as follows: The possible states of each node in the state-coupled model are compared and verified with the set of untouchable bottom line constraints. States that violate any constraint are eliminated, and a preliminary set of feasible states is output. Based on the optional states and their combinations generated by the state coupling model for each node, each state is checked to see if it violates safety constraints, emission constraints, equipment limit constraints, and system stability constraints. During the check, logical judgment or numerical calculation methods are used to map the node state to the corresponding constraint rules. If any node state exceeds the constraint limit, the state combination is removed. All state combinations that do not violate the constraints obtained through this screening process are the preliminary feasible state set.

[0026] For each feasible state marked as a tightly constrained node, output the node constraint identification information; Based on the relative proximity of each node to various constraints, safety margin, load margin, and stability margin indices are calculated. Nodes with safety margins less than a preset threshold or close to the upper limit are marked as tightly constrained nodes. At the same time, the node constraint type, margin value, and influence weight are recorded to generate a node constraint identification information table. This information can be used for state optimization, candidate adjustment direction generation, and executable path planning.

[0027] The initial set of feasible states and constraint identifiers are integrated to generate the final feasible state space; The initially selected feasible state combinations are associated with the corresponding node constraint identification information. The constraint margin and restriction degree of each node of each feasible state are recorded. The output format is a structured data table or file. Each record contains the state combination number, node state value and corresponding constraint identification information. This final feasible state space can be directly used for node change analysis, candidate adjustment direction generation and executable path planning.

[0028] S3: Within the feasible state space, perform a joint analysis of the changes in output, energy consumption, emissions, and equipment load at each node, and calculate the rate of change of the overall gains and losses of the nodes caused by the changes in system state and the rate of change of the safety margin relative to the bottom line constraint.

[0029] The process of jointly analyzing the changes in output, energy consumption, emissions, and equipment load at each node in S3 is as follows: Within the feasible state space, for each feasible state, calculate the node output, energy consumption, emissions, and equipment load indicators, and output a set of node state indicators. Within the feasible state space, the process parameters, material flow rate, energy flow rate, emission data, and equipment operating status of each node in each feasible state combination are read. The node output, energy consumption, emission, and load indicators are calculated using preset calculation formulas or energy consumption models, emission models, and load analysis models. The indicator values ​​of each node form a structured record, and all node indicator combinations are summarized to form a set of node status indicators.

[0030] Analyze the mutual influence, changing trends, and sensitivity between nodes, generate node dependency information, and output a dependency table; Based on the state coupling model and the set of node state indicators, the coupling strength, correlation coefficient and trend indicators between nodes are calculated to determine the impact of a certain node's state change on the output, energy consumption, emissions and load of other nodes. A dependency list is established for each node, recording the affected nodes, the direction of influence and the magnitude of influence, and generating a node dependency information table.

[0031] The node status indicators and information in the dependency table are classified and processed to form a node change classification result; Based on the type and characteristics of the indicators, node changes are classified into different categories, such as increased output, load shift, emission fluctuations, and increased energy consumption. The classification is based on the comparison of the magnitude of the indicator change with a preset threshold or historical average. At the same time, the system impact level of the node change is judged by combining the dependency information. The output is a node change classification table, which is used for candidate adjustment direction screening and optimized path planning.

[0032] Standardize the node status indicators and classification results to output standardized node data; Normalization or standardization operations are performed on all node indicators to enable indicators with different dimensions to be compared under the same measurement scale. For example, output, energy consumption, emissions and load indicators are uniformly mapped to the 0-1 interval or standard scores. At the same time, classification information is encoded into processable numerical labels to form a structured and standardized node dataset. This standardized data can be directly used to generate candidate adjustment directions, calculate the comprehensive gain-loss change rate and safety margin change rate, and ensure that the algorithm calculation results are comparable among different indicators.

[0033] The process in S3 for calculating the rate of change of the overall gains and losses of nodes caused by changes in system state and the rate of change of the safety margin relative to the bottom-line constraints is as follows: Calculate the overall gain and loss change rate of nodes based on standardized node data, and output a gain and loss change result table; Based on the standardized node data obtained in the previous step, the comprehensive gain-loss change rate of each node under each feasible state is calculated. Each node index is assigned a preset weight, such as production weight, energy consumption weight, emission weight, and load weight. The comprehensive gain-loss change rate of the node is calculated by weighted summation or multi-objective evaluation function. The gain-loss change value of each node in the current state relative to the initial state is output. The calculation results of all nodes form a structured gain-loss change result table.

[0034] Calculate the rate of change of node safety margin using the list of bottom line constraints, and output the safety margin result table; Read the baseline constraint parameters, including safety constraints, emission constraints, equipment limit constraints, and system stability constraints. Compare the current state indicators of the nodes with the baseline constraints to calculate the margin change rate, which is the change in the safe distance of the node state relative to the constraint boundary. Output the safety margin result table for each node and record the remaining operating space and change trend of the nodes under the system constraint conditions.

[0035] Integrate the gain / loss change results table with the safety margin results table to output a dual change rate evaluation table; By matching the overall gain / loss change rate of each node with the corresponding safety margin change rate, a node-level bivariate evaluation table is generated. This table is used to simultaneously measure the impact of nodes on the system's economic benefits and safety constraints. The integrated results can be used for candidate adjustment direction screening and priority ranking. The dual rate of change evaluation table is sorted and normalized, and the sorting result table is output. Based on the relative importance of the comprehensive gain / loss change rate and the safety margin change rate, nodes or node combinations are prioritized and sorted. The sorting results are then normalized to map the change rates of different dimensions or ranges to a unified scale, generating standardized sorting results that are easy to use directly in subsequent candidate adjustment direction generation and path planning algorithms.

[0036] S4: Based on the overall gain and loss rate of nodes and the safety margin rate of change, generate a set of multiple state candidate adjustment directions. By comparing the relative changes in the overall gain and loss rate of nodes and the safety margin rate of change in each candidate direction, form an executable path that gradually adjusts from the current state to the target state.

[0037] The process of generating multiple sets of candidate state adjustment directions in S4 is as follows: Multiple candidate adjustment directions are extracted from the sorting result table. Each direction corresponds to a specific combination of node changes, forming a preliminary set of candidate directions. Based on the ranking results, multiple high-priority candidate directions are extracted from the table. Each direction corresponds to a set of specific node state combinations and changes, such as production adjustment, energy consumption regulation, emission optimization, and equipment load changes. During the extraction process, the state coupling model is used to determine the impact of each node combination change on the overall system performance and safety constraints, ensuring that each candidate direction has feasibility and potential optimization effect in the initial stage. This can compress massive state combinations into a limited set of preliminary candidate directions, providing basic data for screening and path generation. At the same time, it ensures that the candidate directions are directly related to the system optimization goals and bottom-line constraints, achieving a close match between technical characteristics and problem-solving.

[0038] After eliminating the initial candidate directions that result in a decrease in overall gains and losses and a reduction in safety margin, a set of candidate directions is obtained after screening. The initial candidate direction set is evaluated, and the changing trends of the overall gain / loss rate and safety margin rate for each direction are calculated. Directions that would lead to a decrease in overall gain / loss or a reduction in safety margin are eliminated to ensure that the remaining directions do not introduce the risk of performance degradation or violation of bottom-line constraints. During the screening process, the interdependencies between nodes are analyzed in conjunction with the coupling model to avoid the negative impact of local optimization on the overall system. The selected candidate direction set retains the optimal executable direction and ensures that each direction has practical significance in terms of system operation safety and output, energy consumption, and emission optimization, thus achieving effective correlation and full disclosure between candidate directions and technical objectives.

[0039] Generate a priority index for the filtered candidate direction set and output a direction priority list; For the selected candidate direction set, a priority index is generated for each candidate direction based on the node comprehensive gain and loss change rate, safety margin change rate, and coupling impact assessment. The priority assessment considers the gain magnitude of comprehensive gains and losses, the degree of improvement of the bottom line constraint margin, and the impact on the load of critical equipment, forming a clear direction priority list. The priority list ensures that the system prioritizes the direction with the greatest overall performance improvement and the least risk when performing path planning.

[0040] Based on the priority list, each direction is mapped to a specific node state combination path, generating multiple sets of state candidate adjustment directions; Based on the priority list, each candidate direction is mapped to a specific node state combination path, which is a continuous state sequence for precise adjustment of the output, energy consumption, emissions, and equipment load of each node. During the mapping process, the state coupling model and the bottom line constraint set are combined to ensure that each step in the path meets the safety constraints, emission constraints, and equipment limit constraints. The generated set of state candidate adjustment directions contains multiple executable paths. Each path clearly defines the node adjustment steps, execution order, and expected effect, providing the DCS system with directly executable adjustment instructions.

[0041] The process of forming an executable path that gradually adjusts from the current state to the target state in S4 is as follows: Select a direction from a set of multiple candidate state adjustment directions, and map the node combination changes corresponding to each direction into a node state adjustment sequence to form a preliminary executable path; Among multiple candidate adjustment directions, the optimal candidate direction is selected as the basis for adjustment by combining the node's overall gain and loss change rate, safety margin change rate and priority index. Each selected direction corresponds to a specific node's state combination change, including production adjustment, energy consumption adjustment, emission optimization and equipment load adjustment. These node combination changes are mapped to a node state adjustment sequence, forming a series of continuous node state change steps, and clarifying the adjustment target of each node at each time step.

[0042] Perform bottom-line constraint verification on the status of each node in the preliminary executable path and generate path verification results; For each node state in the preliminary executable path, rigorous verification is performed based on the set of bottom-line constraints, including safety constraints, emission constraints, equipment limit constraints, and system stability constraints. The verification process determines feasibility by calculating whether the node state exceeds the allowable range, affects the load of critical equipment, or violates coupling constraints. For each step of the operation, the verification results are recorded, including compliance or non-compliance indicators and potential risk indicators.

[0043] The sequence of node states that pass the verification results constitutes the final executable path; The node state sequences verified by the bottom-line constraints in the preliminary executable path are integrated to form the final executable path. The final path clarifies the adjustment operation sequence, magnitude and duration of each node at different time steps, while ensuring that the system meets safety constraints, emission standards and equipment load limits during execution. This path can be directly mapped to DCS control commands to realize the gradual adjustment of the system state from the current state to the target state.

[0044] S5: Map the executable path to adjustment instructions for each product node and equipment unit, and execute the adjustment instructions through DCS to generate status adjustment records.

[0045] The process of generating state adjustment records in S5 is as follows: The state of each node in the executable path and its corresponding adjustment instructions are converted into structured records. Read the state change sequence of each node in the executable path and the corresponding DCS adjustment instructions, and record them in the form of tables or data objects. Each record contains the node number, timestamp, state parameters, and adjustment instruction type and value to form a structured path record for analysis, tracking or backtracking, so as to achieve the traceability and parsing of information.

[0046] Record the output, energy consumption, emissions, and equipment load status of each node during execution, and output a node execution status table; During the execution of adjustment commands by the DCS system, real-time data on output, energy consumption, emissions, and equipment load of each node are collected. The collected time-series data is correlated with the status changes in the path to form an execution status table for each node. Each record includes the node number, parameter value, collection timestamp, and status change type, which is used for system monitoring and operation analysis to ensure that the operating parameters of each node during the execution process can be quantified, stored, and verified.

[0047] The results of the adjustment command execution are integrated with the path verification information to generate a complete status adjustment record; By combining the structured path record with the node execution status table and the bottom-line constraint information verified during the path verification process, a final status adjustment record is formed. The record includes the instructions, actual execution parameters, constraint verification results and anomaly identification information of each node at each step. This record is used for analysis, tracing, operation review and optimization algorithm input, so as to realize closed-loop management of DCS system operation.

[0048] Example 2: As Figure 2 As shown, a DCS intelligent control optimization system for multi-product co-production includes: State modeling module: Based on the process parameters, material flow, energy flow, emissions and equipment status of each node, and combined with the constraints of raw material sharing, equipment coupling and public works, a state coupling model of mutual influence of each node is constructed. Constraint Analysis Module: Utilizes a state-coupled model to generate a set of constraints for safety, emissions, equipment limits, and system stability, filters feasible state spaces, and outputs feasible node states; Direction generation module: Calculates the overall gains and losses and safety margin changes of nodes in the state space of feasible nodes, generates multiple state candidate adjustment directions and sorts them by priority; Path planning module: Based on the set of candidate adjustment directions, it forms an executable path that is gradually adjusted from the current state to the target state, and outputs the path verification results; Instruction execution module: Maps executable paths to node status adjustment instructions, executes them through DCS, and generates complete status adjustment records.

[0049] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A DCS intelligent control optimization method for multi-product joint production, characterized in that, Includes the following steps: Collect process parameters, material flow parameters, energy flow parameters, emission parameters and equipment status parameters corresponding to each product node in the multi-product co-production process, and construct a state coupling model of the mutual influence relationship between the states of each node by combining the raw material sharing relationship, equipment coupling relationship and utility constraint relationship between products. Based on the state coupling model, a set of inviolable bottom line constraints is constructed, including safety constraints, emission constraints, equipment limit constraints and system stability constraints. System states that violate any bottom line constraint are eliminated to form a feasible state space. Within the feasible state space, a joint analysis is performed on the changes in output, energy consumption, emissions, and equipment load at each node to calculate the rate of change of the overall gains and losses of the nodes caused by the changes in system state and the rate of change of the safety margin relative to the bottom line constraint. Based on the overall gain and loss rate of nodes and the safety margin rate of change, multiple sets of candidate adjustment directions for the current state are generated. By comparing the relative changes of the overall gain and loss rate of nodes and the safety margin rate of change in each candidate direction, an executable path is formed to gradually adjust from the current state to the target state. The executable path is mapped to adjustment instructions for each product node and equipment unit, and the adjustment instructions are executed through DCS to generate status adjustment records.

2. The DCS intelligent control optimization method for multi-product joint production according to claim 1, characterized in that, The process of constructing a state coupling model of the mutual influence between the states of each node is as follows: The process parameters, material flow parameters, energy flow parameters, emission parameters, and equipment status parameters collected at each product node are normalized. Calculate the coupling weight matrix between nodes by combining the raw material sharing relationship, equipment coupling relationship and utility constraint relationship between products; A multidimensional state vector is constructed based on the node coupling weight matrix to form a state coupling model in which each node influences the others.

3. The DCS intelligent control optimization method for multi-product joint production according to claim 2, characterized in that, The process of constructing a set of inviolable bottom-line constraints that includes safety constraints, emission constraints, equipment limit constraints, and system stability constraints is as follows: Based on the state coupling model, the parameters of safety constraints, emission constraints, equipment limit constraints and system stability constraints are quantified and represented to form a constraint vector; The constraint vectors are classified and graded to generate a bottom-line constraint matrix based on the node, device, and system levels; Output constraint matrix file, combining the constraint parameters to form a set of untouchable bottom line constraints.

4. The DCS intelligent control optimization method for multi-product joint production according to claim 3, characterized in that, The process of forming a feasible state space is as follows: Compare and verify the possible states of each node in the state-coupled model with the set of untouchable bottom line constraints, eliminate states that violate any constraint, and output a preliminary set of feasible states. For each feasible state marked as a tightly constrained node, output the node constraint identification information; The initial set of feasible states is integrated with the constraint identifiers to generate the final feasible state space.

5. The DCS intelligent control optimization method for multi-product joint production according to claim 4, characterized in that, The process of jointly analyzing the changes in output, energy consumption, emissions, and equipment load at each node is as follows: Within the feasible state space, for each feasible state, calculate the node output, energy consumption, emissions, and equipment load indicators, and output the node state indicator set. Analyze the mutual influence, changing trends, and sensitivity between nodes, generate node dependency information, and output a dependency table; The node status indicators and information in the dependency table are classified and processed to form a node change classification result; The node status indicators and classification results are standardized to output standardized node data.

6. The DCS intelligent control optimization method for multi-product joint production according to claim 5, characterized in that, The process of calculating the rate of change of the total gains and losses of the nodes caused by changes in system state and the rate of change of the safety margin relative to the bottom line constraint is as follows: Calculate the overall gain and loss change rate of nodes based on standardized node data, and output a gain and loss change result table; Calculate the rate of change of node safety margin using the list of bottom line constraints, and output the safety margin result table; Integrate the gain / loss change results table with the safety margin results table to output a dual change rate evaluation table; The dual rate of change evaluation table is sorted and normalized, and the sorting result table is output.

7. The DCS intelligent control optimization method for multi-product joint production according to claim 6, characterized in that, The process of generating multiple sets of candidate state adjustment directions is as follows: Multiple candidate adjustment directions are extracted from the sorting result table. Each direction corresponds to a specific combination of node changes, forming a preliminary set of candidate directions. After eliminating the initial candidate directions that result in a decrease in overall gains and losses and a reduction in safety margin, a set of candidate directions is obtained after screening. Generate a priority index for the filtered candidate direction set and output a direction priority list; Based on the priority list, each direction is mapped to a specific node state combination path, generating multiple sets of candidate adjustment directions for the state.

8. The DCS intelligent control optimization method for multi-product joint production according to claim 7, characterized in that, The process of forming an executable path that gradually adjusts from the current state to the target state is as follows: Select a direction from a set of multiple candidate state adjustment directions, and map the node combination changes corresponding to each direction into a node state adjustment sequence to form a preliminary executable path; Perform bottom-line constraint verification on the status of each node in the preliminary executable path and generate path verification results; The sequence of node states that pass the verification constitutes the final executable path.

9. The DCS intelligent control optimization method for multi-product joint production according to claim 8, characterized in that, The process of generating a status adjustment record is as follows: The state of each node in the executable path and its corresponding adjustment instructions are converted into structured records. Record the output, energy consumption, emissions, and equipment load status of each node during execution, and output a node execution status table; The results of the adjustment command execution are integrated with the path verification information to generate a complete status adjustment record.

10. A DCS intelligent control optimization system for multi-product joint production, applied to the method described in any one of claims 1-9, characterized in that, include: State modeling module: Based on the process parameters, material flow, energy flow, emissions and equipment status of each node, and combined with the constraints of raw material sharing, equipment coupling and public works, a state coupling model of mutual influence of each node is constructed. Constraint Analysis Module: Utilizes a state-coupled model to generate a set of constraints for safety, emissions, equipment limits, and system stability, filters feasible state spaces, and outputs feasible node states; Direction generation module: Calculates the overall gains and losses and safety margin changes of nodes in the state space of feasible nodes, generates multiple state candidate adjustment directions and sorts them by priority; Path planning module: Based on the set of candidate adjustment directions, it forms an executable path that gradually adjusts from the current state to the target state and outputs the path verification results; Instruction execution module: It maps the executable path to node state adjustment instructions, executes them through DCS, and generates a complete state adjustment record.

Images